Shear in slab desig need to be considered or not 2

[tngv752]

I use SAP2000 to design slab, I can shear result in slab.
But I use another Slab software, the output results don't have shear in slab, just moment and reaction.
Someone tell me that slab never fail by shear ?
Seem to be right ?

 

[trainguy]

csd72,

That is one of the truest statements about engineering I've ever seen.

Talk about hitting the nail on the head - Star for you.

tg

 

[demayeng]

I think even though the original post was about shear in slabs, the change to discussing FEA in concrete design is relevant.
I agree with everyone saying you need to know how to check it by hand - it would be foolish to use these tools without knowing how they work. I believe that to increase their use and trust in them, FEA programs should be transparent in their documentation to show what calculation process they use (and if they don't, why they don't consider shear in slabs!!!).

It sounds like people trust their hand calcs over FEA because of accuracy errors in the programs, which is fair enough. But what if these developer errors were fixed and your checks validated the results?

ps, rowingengineer: Slabs now has the option of using the Branson formula to work out Ieff for every finite element (plus you can set different shrinkage strains for either slabs or beams), as well as the approximate method you mentioned..
Thanks for the rapt note - i'll ask the makers of slabs what they say about that!

 

[rowingengineer]

Demayeng,
Sounds like Slabs is starting to make moves in the world. How dose Slabs handle wall/slab connections. Does slabs allow for I eff of the wall stiffness and columns? This would be really starting to get close to the solution. How does it handle the hot spots? I may just have to take anothe look at it.

As for the question, If the developers got it to a point where there were no holes in the program, and it did pattern live loading thus that you didn't need like 20 different load cases. Then I would maybe consider trusting it on a more regular basis. But I should mention that my distrust also comes from the fact that one error in these programs can make a huge difference, And that error can be me. Hand calcs are simple and hard to get wrong.

And one really good thing about building your own calc set is that you get to know things that wouldn'teasly be known if you just followed a program manual.




When in doubt, just take the next small step.

 

[demayeng]

rowingengineer:
Wall slab connections can be pinned or fixed, but they recommend using pinned as it is hard to achieve fixed in real life. Plus they don't have specific wall and column stiffness - probably embedded as an approximate method though. Do you allow for wall stiffness in your hand calcs?

Hot spots - Slabs tells you they are there, but doesn't tell you how to deal with them. Even though stress concentrations do exist (hand calcs generally won't pick them up) concrete seems to behave in a manner that you get local relaxation and they don't often cause major problems (another post topic).

Slabs has covered some of these issues in their documentation, but not all which is pretty frustrating. I picked up a bug in the program early on - they fixed it and sent me a new copy.. another user told me this is common. Seems pretty slapdash.. they need to spend some money on development and make it more professional..

You can do pattern live loading

 

[rowingengineer]

Demayeng,
I normally like to reinforce the edges of slabs running into walls with some reo to stop cracking and ensure the load arrives at the centre of the wall, in my hand clacs I normally allow for some stiffness of wall, generally I make this about 50% of the wall strength. However I would expect most programs would advise that you don't design for moment in the walls.

does the program average hotspots over an area, or just ignore? While I agree hotspots are not all that important to designing a slab because of relaxation ect, They could be fairly important if ignored in a beam design for secondary bending, which I must now ask, how does slabs handle the reinforcement of beams that are large and wide. As i would expect that these would be modelled as plates, not beams, hence you would have compatibility torsion and My? Which brings me to my next question, does slabs handle compatibility torque? Like an edge beam? if so does it iterate it's answer when cracked and uncracked conditions for beams?

I think i could spend many hours trying to find more holes in the program but I guess the truth of the matter is that a program no matter how good can’t replace an engineer.

When in doubt, just take the next small step.

 

[demayeng]

check out the Slabs manual online..

specifically here is the page on wide beams

http://inducta.com.au/help/000068.htm

I don't have an answer for the rest of your questions - at uni and work i haven't learnt about compatibility torsion. However I can't see anything in the program about torsion stiffness..

 

[rowingengineer]

demayeng,
Thanks for the info, have enjoyed the discussion, I will get the boys at inducta to send trial copy, and give the program a working out.
I would recommend taking a look at compatibility and equilibrium torsion in Warner rangan et al. Ch13.1, always good to know things like this for modelling purposes.

When in doubt, just take the next small step.